Quantification of Glargine and Its Metabolites in Human Plasma Using a Hybrid Immunoaffinity Purification and LC-MS/MS Methodology

摘要

Introduction: Glargine is a recombinant human-insulin analog for the treatment of diabetes. Characterization of human pharmacokinetics calls for bioanalytical assays suitable for quantifying glargine and its two active metabolites, M1 and M2, in plasma. Historically, immunoassays have been the primary tool in glargine bioanalysis. Due to inherent specificity limitation, significant interference was observed from other insulin analogs present in diabetes patients. LC/MS techniques provide selectivity based on physicochemical properties, molecular size and charge state, thereby, offering a unique opportunity to address the specificity concerns presented with immunoassays. However, it remains a challenging task to develop a highly sensitive and reproducible LC-MS/MS method for glargine, largely due to the complex biological matrices (e.g. plasma) and technical hurdles associated with peptide analysis. Methods: The assay described herein is based on immunoaffinity capturing and enrichment of intact insulin glargine, M1 and M2 using mouse anti-insulin monoclonal antibody-coated magnetic beads, followed by reversed-phase chromatography coupled with electrospray-MS/MS analysis operated under positive ionization mode. Stable-isotope labeled 6[D_(10)]-glargine and 4[D_(10)]-M1 were used as internal standards. The sample preparation was carried out in a 96-well plate format, using 0.5-mL of plasma. Waters Acquity UPLC BEH300 C4 column (2.1×50-mm, 1.7 μm, 300A) with gradient elution at a flow rate of 0.25 mL/min was employed for LC separation. AB Sciex QTRAP 5500 mass spectrometer was utilized for analyte detection. The LC-MS/MS run time was 3.6 minutes. Preliminary Data: Multiple aspects of the assay were studied to achieve optimal sensitivity and improve assay robustness. Specifically, issues related to analyte nonspecific binding, specificity and reproducibility were addressed and will be discussed. The sample preparation method avoided enzyme digestion and reduction of disulfide bonds. The assay procedure was amenable to automation, and allows for high throughput analysis of clinical samples. The developed method achieved a lower limit of quantitation (LLOQ) at 0.1 ng/mL (～16 pM, or 2.8 microU/mL) for glargine, M1 and M2. The calibration curve range was from 0.1 to 10 ng/ mL. Intra-run precision was within 9.9% (n=5), and accuracy was between 96.0 and 104% for all analytes. Inter-run (n=3) precision was within 15.5%, and accuracy was between 98.5 and 104%. Matrix effect, recovery, analyte stability and interferences from sources including control matrix, potential concomitant medications and anti-drug antibody were assessed. Results suggest the method was accurate, reproducible, and suitable for supporting glargine clinical trials. Novel Aspect: This methodology enables fast and reliable quantification of glargine and metabolites with ultrasensitivity and minimal interferences from other insulin analogs.